Wentao Wang

Informed consent forms are essential for protecting patient rights, but are often difficult to understand, especially in non-English settings where formal language and long, context-dependent sentences hinder comprehension. This study evaluated whether large language models (LLMs) can simplify Chinese language surgical consent forms and whether clinician revision can further improve the output. Official forms from nine hospitals were used to create three versions of each document: the original, an LLM-simplified version, and a clinician-revised version. We assessed text structure, readability, content quality, and layperson comprehension using quantitative metrics and expert ratings. The LLM version improved readability and comprehension but reduced content quality, particularly risk information. Clinician revision restored accuracy while maintaining clarity and achieved the highest comprehension scores. Linear mixed effects modeling confirmed these trends. These findings highlight both the impact of LLM-based simplification and the value of human AI collaboration in patient communication.

Background: Mental health problems among university students are a growing global concern, yet limited counseling resources and inadequate understanding of counseling procedures often delay timely help-seeking. Informed consent forms (ICFs) are essential for safeguarding autonomy and clarifying counseling procedures, but many universities’ counseling ICFs are incomplete, ambiguous, or overly technical. Large language models (LLMs) may offer scalable assistance for improving clarity and accessibility. Objective: This study aimed to evaluate whether LLM-based rewriting could improve the structure, readability, content quality, and comprehensibility of university counseling ICFs, and compared 2 advanced models (ChatGPT [GPT-5] and Grok-4). Methods: We conducted a comparative evaluation of counseling ICFs collected from 33 Chinese universities (original texts) and generated 2 rewritten versions for each ICF using ChatGPT (GPT-5) and Grok-4. A multidimensional framework assessed (1) textual structure and readability, (2) expert-rated content quality from a counselor perspective, and (3) volunteer-rated reading comprehension from a client perspective. Comparisons between original and rewritten texts were performed using Wilcoxon signed rank tests, with linear mixed-effects models used to validate results while accounting for rater variability. Results: Compared with the originals, both LLM-rewritten ICFs showed significant improvements across all evaluated dimensions. The mean Lee-Yang Readability Index decreased from 28.68 (SD 5.69) to 22.39 (SD 2.13) with ChatGPT (GPT-5) and 24.37 (SD 2.32) with Grok-4 (both P <.001), and mean tone friendliness increased from 2.57 (SD 0.29) to 2.67 (SD 0.12) and 2.67 (SD 0.13), respectively. The mean expert-rated content quality improved from 45.33 (SD 8.74) to 52.54 (SD 7.92) and 55.49 (SD 7.81) ( P <.001), driven mainly by higher completeness and specificity of key information. The mean volunteer-rated reading comprehension scores increased from 19.02 (SD 1.32) to 22.33 (SD 0.81) and 22.05 (SD 0.90) ( P <.001), indicating improved clarity, readability, and acceptability. Across structural features, Grok-4 tended to produce longer rewritten forms than the originals, highlighting a potential trade-off between added informational content and document length. Conclusions: In this comparative evaluation of 33 Chinese university counseling ICFs, LLM-based rewriting was associated with improved readability, expert-rated content quality, and volunteer-rated comprehension relative to original forms. These findings suggest that LLMs can support the optimization of counseling documentation; however, implementation should consider practical constraints (eg, document length) and retain human oversight.

Objective This study aims to unveil the role and mechanism of histone deacetylase SIRT1 in modulating fatty acid binding protein 4 (FABP4) stability and its impact on hepatic fibrosis and ferroptosis in non-alcoholic fatty liver disease (NAFLD). Methods A murine NAFLD fibrosis model was induced by a Western diet combined with CCl₄ injection. Mice received adenoviruses carrying oe-SIRT1 or sh-FABP4 three days before modeling. In vitro, NCTC 1469 hepatocytes were exposed to palmitic acid (PA) and transfected with oe-SIRT1 or oe-FABP4. Histopathology was assessed by HE, Masson, and Sirius red staining, and α-SMA by immunohistochemistry. Biochemical assays and molecular analyses measured liver injury markers, ferroptosis-related proteins, oxidative stress indices, and SIRT1/FABP4 expression. Cell viability, death, and lipid deposition were evaluated by CCK-8, LDH release, and Oil Red O staining. Co-IP and CHX chase assays examined SIRT1–FABP4 interaction, acetylation, and protein stability. Results NAFLD mice showed decreased SIRT1 and increased FABP4, with lipid degeneration, collagen deposition, α-SMA upregulation, and aggravated ferroptosis. Either SIRT1 overexpression or FABP4 knockdown alleviated inflammation, fibrosis, and ferroptosis in vivo. In PA-treated hepatocytes, cell viability declined, LDH release and lipid deposition increased, AST/ALT rose, ferroptosis intensified, SIRT1 was suppressed, and FABP4 was upregulated. Enhancing SIRT1 reduced FABP4 expression, attenuated lipid accumulation, and limited ferroptosis. Mechanistically, SIRT1 interacted with FABP4 to decrease its acetylation, thereby lowering FABP4 stability. FABP4 overexpression partially reversed the protective effect of SIRT1, promoting ferroptosis and lipid deposition. Conclusion SIRT1 reduces FABP4 acetylation and stability, thereby suppressing ferroptosis and hepatic fibrosis in NAFLD. Graphical abstract

Background: Microvascular invasion (MVI) critically portends early recurrence and poor survival in intra-hepatic cholangiocarcinoma (ICC) but is ascertainable only after resection. An accurate pre-operative predictor of MVI would aid surgical planning in everyday practice. Objective: To develop and internally validate a blood-based model that integrates serum tumor markers with inflammatory and liver-function indices for pre-operative estimation of MVI in ICC. Methods: This single-center, retrospective, non-interventional cohort included consecutive adults who underwent curative-intent hepatectomy for pathologically proven ICC and completed 3-year follow-up between January 2019 and December 2024. Of 602 patients screened, 450 met eligibility criteria after excluding cases with mixed histology, neoadjuvant therapy, macro-vascular invasion, or missing key data. Pre-operative variables (drawn ≤ 14 days before surgery) comprised CA19-9, CEA, AFP, CA-125, neutrophil-to-lymphocyte ratio (NLR), γ-glutamyl-transferase (γ-GT), albumin, maximum tumor diameter, and lesion multiplicity. Predictors were selected by LASSO, and a shrinkage-adjusted logistic model was internally validated with 1,000-bootstrap resamples. Performance was benchmarked against a tumor-marker-only model and evaluated for clinical utility using decision-curve analysis (DCA). Associations between model-predicted risk, early recurrence (≤ 12 months), and overall survival were explored with Cox regression. Results: Among 450 eligible patients (median age 58 y; 64.2% male), microvascular invasion (MVI) was present in 40.4% (182/450). The final model retained log‑transformed CA 19‑9, CEA, NLR and γ‑GT together with albumin, tumor size and lesion multiplicity. Its apparent and optimism‑corrected AUCs were 0.80 (95% CI 0.76–0.84) and 0.78, respectively—outperforming the tumor‑marker baseline (corrected AUC 0.68; ΔAUC + 0.10). Across clinically relevant thresholds of 12.0–38.0% the extended model delivered consistently higher net benefit than both the baseline and “treat‑all” strategies. Patients assigned to the high‑risk tier (> 29.5% predicted probability; n = 169) showed markedly higher early‑recurrence rates (52.8% vs. 18.6%; adjusted HR 2.9, 95% CI 2.1–4.0) and lower three‑year overall survival (44.3% vs. 72.1%; adjusted HR 2.1, 95% CI 1.5–3.0). Conclusions: A composite model that integrates routine serum tumor markers with inflammatory and liver‑function indices shows strong internal discrimination for pre‑operative estimation of MVI in ICC. Because this is a single‑center, HBV-enriched Chinese cohort, generalizability to other settings and etiologies is uncertain. External, multicenter validation and site‑specific recalibration are warranted before clinical adoption.

Background: Intermediate-stage hepatocellular carcinoma (HCC) following curative liver resection (LR) is associated with high recurrence rates and poor survival outcomes. Current studies have found that transarterial chemoembolization (TACE) can improve overall survival rates and disease-free survival (DFS) in patients with intermediate-stage HCC after surgery. However, the benefits of this treatment are limited. This study aimed to evaluate the benefit of triple adjuvant therapy – TACE combined with antiangiogenic therapy (lenvatinib) plus programmed death-1 inhibitors (TAP) – as an adjuvant treatment for resected intermediate-stage HCC, compared to TACE alone, and to identify patient subgroups most likely to benefit from the TAP regimen. Materials and Methods: We collected data on patients with intermediate-stage HCC who underwent LR from December 2019 to December 2022. DFS was compared between patients receiving TACE and those receiving TAP using propensity score matching. The 2-year recurrence rate in the entire cohort was predicted based on the TACE group, and the association between the predicted and observed recurrences was tested. Outcomes: A total of 571 patients were included in our study, with 102 receiving TAP and 469 receiving TACE. Compared with TACE alone, TAP showed better DFS [hazard ratio: 0.74; 95% confidence interval (CI): 0.56–0.98; P = 0.037], with median: 22.0 months (95% CI: 19.0–24.0) vs. 25.6 months (95% CI: 24.0–40.0). The lines for the TACE and TAP intersected at 37% indicating that patients with a predicted 2-year recurrence risk >37% would significantly benefit from TAP. TAP therapy demonstrated a manageable adverse event (AE) profile, an overall AEs of 79.4%, with grade 1–2 accounting for 65.7%, grade 3 for 11.8%, and grade 4 for 1.9%. Conclusions: TAP therapy demonstrated significant potential as an adjuvant treatment for intermediate-stage HCC following curative resection, offering superior recurrence control and survival benefits compared to TACE alone. Patients with a predicted recurrence risk >37% showed improved DFS outcomes in TAP therapy, suggesting that recurrence risk thresholds could guide tailored treatment decisions in clinical practice. The manageable safety profile of TAP further supports its feasibility in the postoperative setting. Our findings represent a substantial advancement in the field of adjuvant therapy for resected intermediate-stage HCC.

BACKGROUND Mental health problems among university students are a growing global concern, yet limited resources and inadequate understanding of counseling procedures often delay support. Informed consent forms (ICFs) are vital for protecting rights and autonomy, but many are incomplete, ambiguous, or overly technical, and few institutions can effectively optimize them. Large language models (LLMs) offer scalable, low-cost solutions to enhance clarity and accessibility. OBJECTIVE This study aimed to evaluate whether LLM-based optimization could improve the structure, readability, content quality, and comprehensibility of university counseling ICFs, and to compare the performance of two advanced models—ChatGPT-5 and Grok-4. METHODS Counseling ICFs from 33 Chinese universities were collected and optimized using two advanced LLMs, ChatGPT-5 and Grok-4. A multidimensional framework assessed textual structure and readability, content quality from counselors’ perspectives, and comprehension from clients’ perspectives. Evaluations were conducted by mental health professionals and student volunteers. Wilcoxon signed-rank tests and linear mixed-effects models were applied for comparison and validation. RESULTS Compared with the originals, LLM-optimized ICFs demonstrated significant gains across all dimensions. The Lee–Yang readability index decreased from 28.68(5.69) to 22.39(2.13) with ChatGPT-5 and 24.37(2.32) with Grok-4 (both P<.001), while tone friendliness increased from 2.57(0.29) to 2.67(0.12) and 2.67(0.13), respectively. Expert-rated content quality improved from 45.33(8.74) to 52.54(7.92) and 55.49(7.81) (P<.001), primarily through enhanced specificity and existence of key items. Client comprehension scores rose from 19.02(1.32) to 22.33(0.81) and 22.05(0.90) (P<.001), reflecting higher clarity, readability, and acceptability. Linear mixed-effects models confirmed these findings. CONCLUSIONS LLM-based rewriting markedly improved the clarity, completeness, and readability of counseling consent forms. By enhancing linguistic accessibility and professional precision, these models can support clearer communication and stronger counselor–client understanding. For universities with limited counseling resources, integrating LLM-assisted optimization may represent a practical step toward standardized, comprehensible, and client-centered counseling documentation. CLINICALTRIAL Not applicable.

Background & Aims: Hepatocellular carcinoma is the most common type of primary liver cancer and remains a major global health challenge. In resource-limited settings, patients often face barriers such as low screening rates, poor adherence, and limited access to medical information. Despite comprehensive clinical guidelines, issues like inadequate patient education and ineffective communication persist. While large language models show promise in clinical communication and decision support, their performance in hepatocellular carcinoma management has not been systematically evaluated across multiple dimensions. Methods: Ten emerging language models, including general-purpose and medical-domain models, were assessed under prompted and unprompted conditions using a standardized question set covering five key stages: general knowledge, screening, diagnosis, treatment, and follow-up. Accuracy was rated by experts, while semantic consistency, local interpretability, information entropy, and readability were measured computationally. Results: ChatGPT-4o and Grok-3 achieved the highest accuracy (2.62 ± 0.06, 93%; 2.60 ± 0.06, 95%) and interpretability (0.43;0.43). Prompting significantly improved accuracy ( p < 0.001) and interpretability ( p < 0.001) across all models. Semantic consistency declined slightly in most models; information entropy generally increased; readability changes varied. Conclusions: This study presents the first multidimensional evaluation of large language models in hepatocellular carcinoma–related clinical tasks. General-purpose models outperformed some medical models, revealing limitations in domain-specific fine-tuning. Prompt design strongly influenced model performance. Further research should integrate diverse prompt strategies and clinical scenarios to improve the usability of language models in real-world oncology settings. Lay summary: This study evaluated how well-advanced language-based artificial intelligence models can answer clinical questions related to hepatocellular carcinoma. The results showed that some models, especially when guided with structured instructions, provided accurate and understandable responses. These findings suggest that such tools may help improve communication and access to information for both doctors and patients managing liver cancer.

Background The impact of myosteatosis on clinical outcomes following liver transplantation (LT) remains unclear. Methods Articles evaluating the relationship between myosteatosis and the clinical outcomes of LT recipients were comprehensively retrieved from the Embase, PubMed, and Cochrane Library Central databases up to 1 October 2024. Results Thirteen articles involving 3351 cases were included. Myosteatosis was related to increased mortality risk in patients undergoing LT (HR, 1.764; 95% CI, 1.469–2.120; p < 0.01). Patients with myosteatosis had significantly lower rates of 1-year (OR, 0.376; 95% CI, 0.299–0.473; p < 0.01), 3-year (OR, 0.429; 95% CI, 0.353–0.522; p < 0.01), and 5-year (OR, 0.448; 95% CI, 0.343–0.585; p < 0.01) overall survival (OS), higher rates of postoperative complications (OR, 2.482; 95% CI, 1.710–3.601; p < 0.01), 90-day mortality (OR, 4.097; 95% CI, 2.522–6.658; p < 0.01), and early allograft dysfunction (EAD) (OR, 2.197; 95% CI, 1.572–3.071; p < 0.01), and longer intensive care unit (ICU) (SMD, 0.411; 95% CI, 0.122–0.701; p < 0.01) and hospital stays (SMD, 0.682; 95% CI, 0.503–0.860; p < 0.01) than those without myosteatosis. Conclusion Myosteatosis is a potent predictor of poor prognosis and adverse perioperative outcomes in LT recipients. Early detection of myosteatosis can assist in guiding interventions and supportive care to improve the prognosis in the LT population. Furthermore, it is necessary to incorporate myosteatosis into clinical decision-making and selection criteria for LT.

Background: Non-alcoholic steatohepatitis (NASH)-associated hepatocellular carcinoma (HCC) has been emerging a predominant reason for liver transplantation (LT). The complexity of comorbidities in this population increases the possibility of poor transplant outcomes. The purpose of this study was to evaluate the differences in survival after transplantation among patients with NASH HCC and those with non-NASH HCC. Method: We conducted systematic searches of the PubMed, Embase, Web of Science, and Cochrane Library databases. To analyze the data, both fixed and random-effects models were employed to aggregate hazard ratios (HRs) along with 95% confidence intervals (CIs) for recurrence-free survival (RFS) and overall survival (OS) outcomes. This study is registered with PROSPERO as CRD42024578441. Results: A total of seven studies were included in this study. This study revealed that there was no significant difference in OS between liver transplant recipients with NASH HCC and those with non-NASH HCC. The RFS of NASH HCC patients were significantly longer. The HRs were 0.70 (95% CI: 0.51-0.97, P = 0.03) for RFS and 0.88 (95% CI: 0.72-1.07, P = 0.21) for OS, respectively. Conclusion: This study indicates that patients with NASH HCC who undergo LT have comparable OS as those with non-NASH HCC, while NASH HCC was associated with increased RFS. However, further research in randomized trials is necessary to verify these results and address potential selection biases.